In[]:=
ResourceFunction["WolframModel"][{{x,y},{x,z}}{{x,z},{x,w},{y,w},{z,w}},{{1,2},{2,3},{3,4},{2,4}},14,"FinalStatePlot"]
Out[]=

Scalar function on space

Value at every node
[ E.g. put multiple scalar functions to specify “coordinate” ]

Tangent space

Equivalence class of geodesics emanating from a point

Vectors

[[ Basic question: is a vector a single geodesic, or a linear combination of geodesics? ]]
A vector is a geodesic joining two points [on a manifold, it’s OK just to give the points]
Possible vectors from a given point are the possible geodesics, going a certain distance along the geodesic
< Also, could have a vector that is a linear combination of two edges >

Multiplication by scalars

Go further along the vector [multiply by integers]

Adding vectors

Not so obvious, because of parallel transport
[ Go down one geodesic, then from the end point go to another geodesic => non-commutative when there is curvature ]
[ In making a vector space ... normally integer dimensional ]

Vector function on space

To every point in the space, assign a particular geodesic away from that point as the “vector value” at that point

Vector field

Derivation of scalar function [ e.g. directional derivative at each point ]
[ Non-commutative geometry: the “scalar values” do not form a commutative ring; but then one does derivatives anyway ]
< Alternative: at every point, pick an edge >
At every point, associate a number with every edge at that point [ i.e. think of the edges at a point as the basis elements for a vector ]
(When this limits to a manifold, there have to be relations between the numbers on different edges)
[[ With definition of vector as linear combination, derivation is taking differences along each edge ]]

Tensor field

Rank p tensor field: at every point, pick p edges [allowing repeats]
Each collection of p edges is a like basis element in the tensor Tuples[edges, p]

Tensor

Basis is tuples of vectors at a given point
<|1w1,3w2,5w3|>(*vector*)
<|{1,3}w1,{2,4}w2,...|>(*2-tensor*)

Dual vectors

Assume a vector is defined by a number associated with each outgoing edge
[[Linear function from vectors to scalars]] f[new vector] : like a projection onto the existing vector
< cotangent space >

Volume form

All the outgoing edges around a given point

(Dual to) p-Form

Basis is all collections of all-but-p of the outgoing edges

Tensor product

I.e. outer product, where one forms all combinations of the keys

Wedge product

u⋀v : take all the edges associated with u and concatenate them with v
When the resulting component is {1,3}, the weight is the negative of when it’s {3,1}

Totally antisymmetric tensor [?]

<|{even}1,{odd}-1,XXX|>
<|{1,2,3,4,5}1|>

Rotation groups